The present invention relates to a solid state imaging device, and more particularly to a solid state imaging device to be used for a digital still camera, a video camera or the like.
In a solid state imaging device, signal charges accumulated in an imaging area are picked up as electric signals from an output unit via vertical transfer units and a horizontal transfer unit, and the picked-up electric signals are recorded in a video apparatus or directly transferred to a television or the like positioned at a reception side to be reproduced.
For a solid state imaging device in a related art adopting a charge transfer method of an interlace scan-interline transfer (IS-IT) type, a technique of thinning pixels in a vertical direction has been proposed in order to improve a read speed (hereinafter called a frame rate). In one example of pixel thinning, in order to realize a frame rate of 30 frames/sec in a solid state imaging device whose frame rate for outputting all pixels of 1280×960 is 7.5 frames/sec, only one pixel among four pixels in the vertical direction is read, i.e., compression is performed at a compression ratio of 1/4 which is a ratio of reading pixels in the vertical direction, to thereby improve the frame rate.
FIG. 6 is a schematic diagram of an electrode structure illustrating pixel thinning in the vertical direction. A drive pulse indicated by a symbol H in FIG. 7 is applied to a vertical electrode 107, a drive pulse indicated by a symbol I in FIG. 7 is applied to a vertical electrode 108, a drive pulse indicated by a symbol J in FIG. 7 is applied to a vertical electrode (2) 109, a drive pulse indicated by a symbol K in FIG. 7 is applied to a vertical electrode 110, a drive pulse indicated by a symbol L in FIG. 7 is applied to a vertical electrode 111, and a drive pulse indicated by a symbol M in FIG. 7 is applied to a vertical electrode (4) 112. In this manner, as shown in FIG. 8, signal charges accumulated in photo diodes indicated by reference numerals 101 to 110 among photo diodes 101 to 140 are read to vertical transfer units and the read charges are transferred from the vertical transfer units to a horizontal transfer unit. Specifically, signal charges accumulated in the photo diodes indicated by reference numerals 106 to 110 are read to the vertical transfer units, by applying a read voltage, i.e., a read voltage (1) 113 for reading signal charges accumulated in the photo diodes to the vertical transfer units, to a drive pulse, i.e., a vertical transfer clock pulse indicated by the symbol K in FIG. 7 among the drive pulses for driving the vertical transfer units indicated by the symbols H, I, J, K, L and M. The read charges are transferred by two lines by the vertical transfer clock pulses. Thereafter, signal charges accumulated in the photo diodes indicated by the symbols 101 to 105 are read to the vertical transfer units, by applying a read voltage (2) 114 to the vertical transfer clock pulse indicated by the symbol H in FIG. 7. The signal charges accumulated in the photodiodes indicated by reference numerals 101 to 110 and read to the vertical transfer units are sequentially transferred to the horizontal transfer unit by the vertical transfer clock pulses. The charges transferred to the horizontal transfer unit are sequentially transferred from the horizontal transfer unit to the output unit by horizontal transfer clock pulses indicated by symbols N and O in FIG. 7, to be output from the output unit and realize compression in the vertical direction at a compression ratio of 1/4. In FIG. 6, “G”, “R” and “B” symbols represent “green”, “red” and “blue” of color filers, respectively, and a symbol P in FIG. 7 represents a horizontal blanking interval. FIG. 8 shows the positions of signal charges read at a timing indicated by a symbol Q in FIG. 7.
Although compression in the vertical direction can be performed in the manner described above, compression in the horizontal direction is difficult. Therefore, for example, as pixels of 1280×960 are compressed in the vertical direction at a compression ratio of 1/4, pixels of 1280×240 are output from the output unit. An aspect ratio for TV cannot be maintained, so that a balance between vertical and horizontal directions is degraded. There arises therefore a fact of a necessity of maintaining an aspect ratio of 4:3 by performing compression at a compression ratio of 1/4 in a post process.
In order to address this fact, a technique (for example, refer to Japanese Patent Application Publication No. 2002-112122) has been proposed. In the technique, a predetermined number, two or more, of adjacent vertical transfer units are collectively used as one group, each group is made in one-to-one correspondence with a unit transfer bit of the horizontal transfer unit, and charges are transferred from the vertical transfer unit to the unit transfer bit (hereinafter called a corresponding unit transfer bit) of the horizontal transfer unit corresponding to the group to which the vertical transfer unit belongs, to thereby conduct compression in the horizontal transfer unit in the horizontal direction.